Completed
- Addressing Over-Personalization & Filter Bubbles
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
How Netflix Built a Single Model for Search and Recommendations
Automatically move to the next video in the Classroom when playback concludes
- 1 - The Challenge: Scaling for 300M+ Users
- 2 - The Two-Stage Ranking Framework
- 3 - Introducing UniCoRn: One Model, Four Use Cases
- 4 - System Considerations: Latency, Throughput, and SLAs
- 5 - Building a User Foundation Model The "Harry Potter" of ML
- 6 - Tokenizing User History: Titles vs. Words
- 7 - Results: The Impact of Personalization Magic
- 8 - Addressing Over-Personalization & Filter Bubbles
- 9 - Q&A: Fine-tuning, Cold Starts, and Multi-modal Data
